A Synonymous Change in the Influenza A Virus Neuraminidase Gene Interferes with PCR-Based Subtyping and Oseltamivir Resistance Mutation Detection

LETTER

Rapid detection of the emergence of oseltamivir-resistant influenza A virus is critically important for both public health surveillance and the selection of appropriate antiviral therapy. The most common oseltamivir-resistant influenza A (H1N1) 2009 virus genotype is a cytosine-to-thymine mutation at nucleotide 823 of the neuraminidase (NA) gene (823C>T), resulting in a histidine-to-tyrosine mutation at amino acid 275 (His275Tyr, N1 nomenclature). The close correlation of this mutation with oseltamivir resistance has allowed the development of numerous nucleic acid amplification-based methods for its detection, including a duplex real-time, reverse transcriptase PCR (rRT-PCR) with allele-specific hydrolysis probes that we implemented in our laboratory (Table 1). Here we report the identification of a thymine-to-cytosine change at nucleotide 822 (822T>C) that does not alter the amino acid (Tyr274) but does interfere with probe binding in our laboratory-developed assay. Review of the literature suggests that this variant may necessitate the redesign of many published assays for influenza A (H1N1) 2009 virus oseltamivir resistance mutation detection.

Table 1.

Influenza A (H1N1) 2009 virus oseltamivir resistance assays^a

Assay and primer or probe name	Sequence (5′ to 3′)b	Label (5′/3′)	Reference(s)
rRT-PCR
NA Forward	CCAGTTATCCCTGCACACACAT		This study and 2
NA Reverse	CAGTCGAAATGAATGCCCCTAA
NA-WT probe	AGCATTCCTCATAGTGATA	CalFluor560/BHQ-1+
NA-tamifluR probe	AGGAGCATTCCTCATAGTAAT	FAM/BHQ-1+
NA-WT probe V2	AGCATTCCTCATAGTGRTA	CalFluor610/BHQ-2+
NA-tamifluR probe V2	AGGAGCATTCCTCATAGTART	FAM/BHQ-1+
panN1-H275-sense	CAGTCGAAATGAATGCCCCTAA		8
panN1-H275-antisense	TGCACACACATGTGATTTCACTAG
panN1-275H-probe	TTATCACTATGAGGAATGc	FAM/BHQ-1
panN1-275Y-probe	TTATTACTATGAGGAATGc	Dragonfly/BHQ-2
Forward primer	GGGGAGTGGCTGTGTTA		4
Reverse primer	AGGGCGTGGATTGTCTCC
Sensitive probe	TCCTCATAGTGRTAATT	FAM/NFQ
Resistant probe	TCCTCATAGTARTAATT	FAM/NFQ
For-mutant	AGGATAACAGGAGCATCCTAGCA		6
For-wild-type	GGATAACAGGAGCATCCTAACG
Rev-common	ATGGACAGGCCTCATACAARATCTT
Probe-common	TTGACTATCTTTCCCTTTTCT	FAM/MGBNFQ
SwNA-782	GAAAGATAGTCAAATCAGTCGAAATGAA		3
SwNA-887	CAGTTATCCCTGCACACACATGT
SwNA-815-WT	CTAATTATCACTATGAGGAAT	FAM/MGBNFQ
SwNA-815-Mut	ATTCCTCATAGTAATAATTAG	VIC/MGBNFQ
H275-sense	CAGTCGAAATGAATGCCCCTAA		1
H275-antisense	TGCACACACATGTGATTTCACTAG
H275Y-probe	ATTACTATGAGGAATGd	Cy5/BBQ
H1N1pdm-His275Tyr For	TGGACAGGCCTCATACAAGA		7
H1N1pdm-His275Tyr Rev	TGGACAGGCCTCATACAAGA
H1N1pdm-His275	CCTAATTATCACTe	Eclipse/FAM
H1N1pdm-Tyr275	ATTACTATGAGGAe	Eclipse/ROX
SOIV NA 781F	GGAAAGATAGTCAAATCAGTCGAAATG		10
SOIV NA 878R	CTGCACACACATGTGATTTCACTAG
SOIV Osel SEN	CCTCATAGTGATAATTA	FAM/MGBNFQ
SOIV Osel RES	CCTCATAGTAATAATTAG	VIC/MGBNFQ
RT- PCR/RFLP
NA Forward	AATCAGTCGAAATGAATGCCCCTAATGAT		5
NA Reverse	CGATACTGGACCACAACTGC
RT-PCR/RCA
NA Forward	GGACCAAGTGATGGACA		9
NA Reverse	CCACAACTGCCTGTCTT
H1N1 swine 274Y RCA probe	RTAATTAGGGGCATTCATTTCGGATCATGCTTCTTCGGTGCCCATGAGGTGCGGATAGCTCGCGCAGACACGATAGTCTAAGGAGCATTCCTCATARTA
H1N1 swine 274H RCA probe	RTAATTAGGGGCATTCATTTCGGATCATGCTTCTTCGGTGCCCATCCTAGATCAGACGTTCCTGTCCGCGCAGACACGATAGTCTAGGAGCATTCCTCATARTG
RCA primer 1	ATGGGCACCGAAGAAGCA
RCA primer 2	CGCGCAGACACGATA

^aAbbreviations: rRT-PCR, real-time, reverse transcriptase PCR; RT-PCR/RFLP, reverse transcriptase PCR/restriction fragment length polymorphism; RT-PCR/RCA, reverse transcriptase PCR/rolling circle amplification; FAM, 6-carboxyfluorescein; WT, wild type; BHQ, black hole quencher; NFQ, nonfluorescent quencher; MGBNFQ, minor groove binder, nonfluorescent quencher; BBQ, blackberry quencher; ROX, 5-carboxy-X-rhodamine.

^bBold underlining indicates position 823; bold italic indicates position 822.

^cProbe uses locked nucleic acids; see reference 8 for details.

^dProbe uses locked nucleic acids; see reference 1 for details.

^eProbe uses an RNA base; see reference 7 for details.

Between 18 October 2010 and 12 April 2011, we tested 133 influenza A virus-positive respiratory specimens for the Tyr275 oseltamivir resistance mutation. Wild-type 2009 (H1N1) virus NA RNA was detected in 71, while mutant RNA was found in one, a sample from a pediatric heart transplant recipient on oseltamivir. The remaining 61 samples did not give a signal in the assay, but amplifiable influenza A virus matrix RNA was confirmed using the CDC protocol. These specimens were presumed to contain influenza A (H3N2) virus and were sent to the Santa Clara County Public Health Laboratory or tested in-house using commercial reagents (Gen-Probe Prodesse ProFast+) for type confirmation. H3 RNA was detected in 55, while 2009 H1 RNA was unexpectedly detected in six.

After ruling out specimen mix-up and technical error, the most likely explanation for these typing results was an issue with our assay. Agarose gel electrophoresis of the rRT-PCR mixtures demonstrated strong bands at the expected size (92 bp), indicating that the primers were functional and that sequences targeted by the probes were likely mismatched. Sanger sequencing of the amplicons using the assay primers revealed a single thymine-to-cytosine change at nucleotide 822 (822T>C; Tyr274Tyr) in all six of the discrepant samples. Of the 4,931 human 2009 (H1N1) virus neuraminidase sequences in the NCBI database (accessed 30 May 2011), 32 contained the same single nucleotide polymorphism (32/4,931; 0.65%). These sequences were identified from viruses collected throughout 2009 and 2010 in Portugal, Spain, Poland, Iran, Kazakhstan, Canada, and the United States, suggesting the random appearance and widespread distribution of this variant.

As this base change is directly adjacent to the mutation that confers oseltamivir resistance, we evaluated the sequences of primers and probes in published assays to determine whether those assays might also be affected. We identified nine additional H275Y assays in the literature (1, 3–10) (Table 1). Of these tests, only three took this variant into account (4, 6, 9). Given that numerous strategies were employed to increase target specificity, including minor groove binders, locked nucleic acids, and RNA bases, the other assays would likely be unable to tolerate this additional mismatch, and mutant or wild-type sequence containing the variant may not be detected. In addition, an erroneous subtype could be assigned if the assay was not used in concert with subtyping based on another region of the influenza A virus genome.

Introduction of a degenerate base in both probes in our laboratory-developed assay (Table 1) allowed detection of wild-type 2009 (H1N1) virus NA RNA in all six samples carrying this neuraminidase sequence variant. These observations reinforce the importance of regular evaluation of assay target sequences against newly available database sequences, particularly for RNA viruses like influenza A virus that have high mutation rates.